Visual Working Memory (VWM) for objects and their locations underlies the ability to represent, learn, and reason about the visual environment. Tracking the development of VWM for 'object- location bindings' is required to understand scene memory, visual search, and the development of object cognition in general. While it is known that performance in VWM tasks increases with development, it is still undetermined what mechanism actually develops. Our paradigm will probe the attention/VWM interface, specifying the relative contribution of the development of VWM capacity per se, versus the development of focused attention (recent advances in pupillometry allow for the online measurement of this heretofore neglected factor - a factor capable of modulating the effective capacity of VWM from zero to maximal level). Our preliminary work shows that both processes undergo rapid maturation and they both are powerful contributors to overall VWM performance. Here we introduce Delayed Match Retrieval (DMR), a formal, extensible paradigm that requires participants to actively apply 'what is where' information. Our paradigm has several innovations to address challenges that may have stymied previous developmental research in this area: 1) VWM tests can be used with both infants and toddlers, enabling the tracking of developmental trends, 2) stimuli are calibrated to allow for fair comparisons between age groups and features, 3) capacity is assessed with standard measures (Cowan's K), and 4) pupillometry is used to measure the modulatory effect of focused attention on VWM on a trial-by-trial basis. Put together, we will track the development of the attention/VWM interface across the first two years of life cross-sectionally (in 10-, 14- and 22-month-olds) and within individuals in a longitudinal study (given that individual differences in VWM capacity have a remarkable stability in older children and performance in VWM tasks highly correlate with non-verbal IQ in both children and adults). VWM development is a competition between the maturation of attention and underlying capacity; we predict capacity gains will outpace gains in the control of focused attention. We also predict that developmental trends for individuals in our longitudinal study will be stable and have a consistent relationship to the group trend. Our paradigm is also well-suited for systematic comparisons between typically and atypically developing children, or infants born with a high genetic risk for neurodevelopmental disorders (ADHD, OCD). Beyond this, we anticipate a broader impact on translational research, as VWM is impaired in various conditions (e.g. Mild Cognitive Impairment, Williams Syndrome), while attentional processes are affected in others (e.g. ASD).